We propose to test the hypothesis that the medical record both for inpatient, and outpatients at Boston's Beth Israel Hospital can be rendered entirely paperless. While this goal may not be completed within the proposed 3 year project period, we believe that the success that we have had with out computerized Online Medical Record (OMR) used in an outpatient setting, and the extensive inpatients infrastructure now in place portend a reasonable probability of success. During the 3 year project period we plan to make sufficient progress to make completion compelling. Computer-based medical record systems have been in evolution for over two decades. These systems have evolved along two separate pathways - one for hospitalized patient care focused primarily on results reporting and one for non-hospitalized patient care focused more on organizing clinical narrative and periodic care initiative. However, these efforts have neither focused on the interaction of the clinician with the computer- based medical record nor on the sharing of the patient's record across disciplines, across traditional inpatient-outpatient boundaries, and across a geographically dispersed community. WE have designed an online medical record for an office-based general medical practice. Physicians in this practice now routinely use this system for the care of general medical patients. The system is well liked and heavily used. The system encourages clinicians to enter data directly. In developing OMR, we have included data elements and program structure that we have found to be necessary and sufficient to virtually eliminate the use of paper for charting by most providers at HCA. We also have shown that this online medical record is an effective vehicle for delivering clinical practice guidelines. We propose to transform this online medical record into a community-based medical record that serves medical and surgical specialty interests, and office-based and hospital-based needs. This shared record will promote collaboration and communication among care providers, and will serve as a test-bed for developing a structural and functional model of an online medical record. We will focus on exploring the structure and function of the problem and medication lists as essential elements of the medical record. The problem list is the keystone for our understanding of the online medical record. It facilitates navigation of clinical narrative, provides potential trigger points for clinical guidelines, and along with the medication list, provides a valuable patient summary. We will use our heavily used and highly interactive problem lists as a test-bed to help empirically derive a problem list vocabulary. We will also focus on methods to monitor the confidentiality of online patient information and will develop metrics to evaluate confidentiality.